GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Is dense-pack cellulose a good option in the roof slope of a 1960s cape?

4xL7Ho7NZM | Posted in Energy Efficiency and Durability on

I’m doing a lot of work in a town on Long Island, NY (zone 4) that has many cape-style houses. Typically there is poorly installed insulation in the kneewall, kneewall floor and the sloped ceiling above the second floor.

We are proposing:
1. Removing any existing fiberglass batt insulation
2. Installing 1 or 2 inch polyiso board under the roof rafters to create a void.
3. Blocking down at the soffit to eliminate air migration into the roof slope
4. Blocking at the top of the ceiling slope to stop the cellulose and allow us to dense pack
5. Dense packing the void so that the dense pack would extend from the top of the first floor wall to the top of the sloped ceiling on the second floor.
5. The attic above the 2nd floor, when accessible would be air sealed and have loose fill cellulose added.

I like this for a few of reasons:
– The kneewall storage becomes part of the conditioned space. This eliminates the need to air seal or try to insulate the kneewall
– There is a continuour layed of dense-pack cellulose following essentially the entire roof line of the 2nd floor.
– It address the ceiling slope, which would be hard to address another way with out removing sheetrock.

The concerns that have been raised:
– Won’t the cellulose become a huge moldy mess if there is a roof failure.
– it is difficult to create a perfect air seal with cellulose, so could you have a condensation issue against the roof sheathing
– This is not a code compliant structure unless you:
a. put a vapor impermeable insulation of at least R-15 against the roof sheathing. closed cell foam might take care of this at least in the kneewall slope, but would not help where there is already ceiling.
b. leave the underside of the structure with a vaper permeable surface so the structure could dry if it becomes wet.

I’ve been told that my proposal is compliant with the 2009 IEC, but of course no one seems to actually be able to put their hands on one – I guess they don’t want to pay the $109 either.

Any thoughts of comments on this are appreciated.

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. GBA Editor
    Martin Holladay | | #1

    You are correct that the best way to insulate a Cape is to follow the roof slope from the top of the perimeter wall all the way to the ridge. Insulating the kneewall doesn't work.

    Insulation contractors in Vermont and other cold climates have been dense-packing cathedral ceilings, with and without ventilation channels, for decades. Advocates of dense-packed insulation swear by the method -- and, if it's done right (with an air-sealed ceiling), I'm sure the technique works.

    Of course, one of the problems is that you need to provide a deep enough cavity that you can install a code-compliant layer of cellulose. In Zone 4, you want a minimum of R-38 ceiling insulation -- which means about 10 1/2 inches of dense-packed cellulose.

    I'm not going to rule on whether your proposed insulation of dense-packed cellulose is in compliance with the local building code. That ruling needs to be made by your local building official, who should be consulted before you settle on a plan.

    Many building scientists prefer to see rigid foam insulation installed above the roof sheathing in an installation like yours -- enough rigid foam to keep the roof sheathing above the dew point during the winter. Failing that, they like to see closed-cell spray foam installed on the underside of the roof sheathing.

    If you aren't going to install any foam, I think it's imperitive that you include a soffit-to-ridge ventilation channel in each rafter bay.

  2. 4xL7Ho7NZM | | #2


    Thanks for the feedback. SInce these are all existing buildings, I'm stuck with the roof rafters that they give me. In these small houses they are generally 2x6 and sometimes 2x8. Well short of an R-38, but would be a heck of a lot better than the R-13 fiberglass I'm generally pulling out.

    How would you recomend doing a ventilation above the ceiling? The kneewall area is easy enough, but getting something up into those ceiling channels without dropping the sheetrock seems hard.

  3. GBA Editor
    Martin Holladay | | #3

    I don't recommend doing a job halfway. If you're going to insulate, you need to find a way to achieve R-38 or better. If you're not installing new roofing (and therefore limited to working from the interior), I'd recommend closed-cell spray polyurethane foam.

    If your foam contractor needs to remove parts of the ceiling for access, then it's time to poke some holes in the ceiling. Ceilings are easy to patch.

  4. Michael Blasnik | | #4


    I don't think dense paxck cellulose would really be just halfway -- maybe 90% of the way.... I did a quick calculation and estimate that the payback from switching to spray foam from cellulose for a 2x8 rafter cavity would be about 200 years.

    I assumed R-26 for the cellulose cavity and R-42 for spray foam, cellulose at $1.50/sqft and foam at $7/sqft (which would be quite cheap if you actually had to open up the ceiling). I estimate the foam would provide extra energy savings of about 20 therms per year for 1000 sqft of slopes giving you savings of maybe $20-$30/year compared to an extra cost of $5500.

    I think you'd get a better energy payback by replacing the plasma TV with a new LED backlit one ;}

  5. Riversong | | #5


    Per your concern about "a huge moldy mess if there is a roof failure": cellulose with borates is one of the most mold-resistant of insulations. Cellulose can absorb and release up to 30% of its weight in water, which is why it's also one of the best moisture buffers. Beyond that level, it will become paper maché and have to be replaced. But at least it will reveal a roof leak which spray foam will not, concentrating the moisture in the roof sheathing and framing until they mold and rot.


    My experience has been very different than Robert's on the roof leak issue. I've had two leaks in damp spray cellulose and, while the cellulose didn't mold, the OSB and framing that was in contact with the wet cellulose certainly did. The cellulose also accumulated a great deal of water before the sheetrock got wet and tipped the home owner off about the slow leaks.

    I've also had leaks in open cell spray foam and in every case the water came through and wet the sheetrock very quickly and there was no rot due to the fact that very little water accumulated in the foam, most of it ran right thorough to the sheetrock.

    Just my personal experience here, but run a drip in your sink over a block of spray foam one night and a fist full of cellulose the next (and a fist full of JM Spider the third) and tell me which is likely to make "a huge moldy mess in the case of a roof failure." The cellulose absorbs far more than 30% of it's weight very very quickly.

    Robert, we've been over this before and I'm not going to debate you on this, I respect the depth of your knowledge very much, I just have a different experience with this one aspect of this issue and I can't let slide your claim that cellulose is the best insulation in the event of a roof leak. It's a great insulation so long as you keep it dry. Open cell foam and Spider are much more tolerant of occasional wetting in my experience.

  7. 4xL7Ho7NZM | | #7


    The issue with foam in this application is the difficulty in applying it to the closed ceiling without dropping the sheetrock. Dense pack fiberglass seems like it might work well, but it's not a product that I've used. We are using the Accu1 9218 currently for Cellulose. It is rated to be able to use either rockwool or fiberglass as well. Do you or anyone have any experience on how the experience might be different? How about the cost of material? Last, are the different fiberglass products, basically the same or are there differences in quality, cost or application that I should be aware of?


    The individual fibers of the JM Spider fiberglass are significantly smaller in diameter than the fibers of conventional fiberglass. I think its 2 microns as opposed to 4 to 5 microns, so you can get more fibers per cubic inch and therefore more air spaces between the fibers and greater resistance to air flow leading to similar performance to Cellulose and open cell foam. they also seem to ball up like little cotton balls and are somewhat less irritating to the skin due to the lesser stiffness. Obviously you don't want to be in a room when the stufff is being installed without a breathing mask.

    I always hire pros to do my insulation so I can't say how it would respond to the different equipment. I assume JM requires factory approved training for it's installers but it looks like a conventional blown-in-batt installation where you cover the wall with a scrim and punch holes in it and pack in the dry fiberglass. They also offer a spray -in-the air approach that wouldn't work for your application and doesn't seem to save much money in our market.

  9. weatherizemaine | | #9

    I have insulated 124 homes wiht cellulose insulation in slopes and attic flates and spray foam or xps in crawls and basements. I densepack slopes everyday with no ventilation. I am told to do this by a maine housing home inspecter. In story and a half home behind the kneewalls we will put 1" tuff-R cellotex up. We tap all seams and foam the edges to ensure an air seal. Seal all sheetrock and be sure the cellulose is densepacked. Cant do it with a rental machine. Attach a piece of 2" pvc to your 2" insulation hose and slide it all the way from the attic to the eave behind the sheetrock or plater and lath and celotex. Turn the machine on and almost the whole slope will fill with your pvc still at the bottom. Plug the top of the slop around your hose with fiberglass to get a densepack all the way to the top of the slope. When the hose clogs pull it out about 6" till it clogs again. As long as you have atleast a 2x6 slope this will work for you just fine.

  10. weatherizemaine | | #10

    sorry for the typos

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |